Marijuana smoking has biologic plausibility as suspected rish factor for these smoking-related cancers. However, the vast majority of published reports on marijuana-associated cancer are case series without comparative data from an adequate control group. The specific aims of this study are: To estimate whether and how strongly marijuana use might signal an increased risk for cancer of the head, neck, or lung before and after adjusting for use of tobacco and alcohol. To investigate the hypothesis that marijuana smoking co-occurs and clusters with cancer risk factors in specific (e.g., asbestos exposure), and general chronic disease risk factors in general(e.g., sedentary lifestyle, high-fat diet).